The OECD says yes; how do we respond?

A recently released OECD report concludes that economic inequality hurts economic growth, and has particularly done so for New Zealand. Some of our responses were plain bizarre. Either the non-economic commentators had not understood the issue or had not read the report.

Even the Minister of Finance, Bill English, said the research was from a ‘bunch of econometricians doing their magic’. As it happens, English’s degree is in literature, and he has as much authority to judge econometricians as I have to judge poetry (much of which is magical to me).

Almost certainly, he is referring – somewhat loosely – to the report on the research paper which the Treasury prepared for him. Because economics is a disciplined profession, other economists can infer what the Treasury analysis says. (One is not so confident in predicting their policy recommendations.)

The OECD report is based on a working paper Trends in Income Inequality and its Impact on Economic Growth prepared by Federico Cingano. It opens with the caveat ‘OECD Working Papers should not be reported as representing the official views of the OECD or of its member countries. The opinions expressed and arguments employed are those of the author(s).’ As in the case of our Treasury and the RBNZ, I have no trouble with that policy; it allows the OECD to publish research findings for public discussion without having the full imprimatur and political direction from the top.

Yes, the report is based on econometrics which involves non-experimental statistical inference; one of the hardest areas of statistics. As an introduction, consider the issue of whether unemployment causes human distress.

An experimental scientist would take a sample – it would have to be a very large sample – split it randomly into two groups, make one of the groups unemployed, and see whether its members suffered more physical sickness, matrimonial breakdown, criminal or deviant behaviour, mental breakdown, suicide and death. Not practical? Then do we just depend on opinion?

Instead we take what data there is available and analyse it as best we can. Some years ago I looked at over a hundred studies – only a few from New Zealand – which almost unanimously found a correlation between individual unemployment and personal distress. In my judgement almost every one of the papers had some technical flaw, but collectively they were so overwhelming that I think we can safely conclude that unemployment is bad for an individual – caveats, of course.(I also looked at whether inflation is bad for individuals; there is no such body of statistical evidence.)

My general conclusion, though, was never to trust a single piece of research based on non-experimental statistical inference EVEN when I like its conclusion. It may suggest further research directions, it may be indicative, it may even be helpful – but it is not definitive. I doubt that Dr Cingano would particularly disagree.

However his single piece of econometrics contributes to a major OECD (and international) research program containing many other studies. It has led the OECD to some general policy conclusions, not all of what were conveyed in the New Zealand reports.

So here they are (with the empirical findings removed for space; their embolding)
1. What matters most for growth are families with lower incomes slipping behind. This negative effect of inequality on growth is determined not just by the poorest income decile but actually by the bottom 40% of income earners.
2. This is because inter alia people from disadvantaged social backgrounds underinvest in their education.
3. Tackling inequality through tax and transfer policies does not harm growth, provided these policies are well designed and implemented.
4. In particular, redistribution efforts should focus on families with children and youth, as this is where key decisions on human capital investment are made and should promote skills development and learning across people’s lives.

So, if we want to accelerate economic growth and/or ameliorate economic inequality what might we take from the OECD conclusions?

The last two points favour carefully designed tax and transfers policies focussed on families with children and youth. That is where the foundations of the human capital investment which promotes economic growth and wellbeing are laid.

That does not mean we should ignore the government’s role in infrastructure, innovation, trade policy, directly upgrading workforce skills and so on. But if we don’t get the family foundations right they will not be nearly as effective.

I don’t think OECD economists, nor Treasury economists, would be uncomfortable with this conclusion (although they may well add a number of other things). Our politicians are more likely to put the tax and transfer revisions into the too-hard basket. That is political magic.

Comments (4)

by Lee Churchman on December 15, 2014
Lee Churchman

A better argument is from John Rawls: it is morally wrong to allow the worst off people in society to be worse off than they would be under some other possible distribution. Treating people as a means to the end of economic growth is not treating them as ends in themselves. 

by BeShakey on December 16, 2014

That's a much worse argument. Rawls and Kant are both controversial - a lot of people accept them, but many don't (I know more about Kant than Rawls and I'm confident in saying most moral philosophers in Australasia aren't Kantians). Worse, it's an abstract academic position that won't have any traction with the public.

From a practical point of view, it's a much better argument (in the sense of being more likely to achieve change) to say that inequality is a handbrake on growth and there are effective mechanisms to address inequality that won't harm growth.

by Lee Churchman on December 17, 2014
Lee Churchman

Worse, it's an abstract academic position that won't have any traction with the public.

As is the utilitarianism implicit in the pro-growth view. You will find that regular people are not unabashed utilitarians either – it is, as you said of the others, a controversial view, so your complaint finds too many targets to be effective. 

When people ask if they approve of pro-growth policies they don't tend to be very clear about what they are assenting to. It's not hard to find similarly vague and appealing slogans that exemplify Kantian or Rawlsian positions (economic rights, guaranteed minimum incomes, etc.) if you wanted to persuade voters. Democracies don't really work that way most of the time anyway. 


by Brian Easton on December 22, 2014
Brian Easton

I am greatly influenced by Rawls too, Lee Churchman, but his moral principle does not obviously tell us how to organise society to implement it. To do so, we need an understanding of how society and the economy works. That is what the column is about.

Suppose reducing inequality gave the poor a temporary increase in their well-being but in the long run reduced it (because the economy did not work as well). That was a widely held view in the past. What the research I was reporting suggests is that the view need not be correct. It also sets out a mechanism (via human capital) which explains why it might not be. 

Post new comment

You must be logged in to post a comment.